Selective Extraction of 7-aminodeacetoxycephalosporanic acid and Cephalexin antibiotics with Aliquat-336

1. Available on line www.jocpr.com Journal of Chemical and Pharmaceutical Research __________________________________________________ J. Chem. Pharm. Res., 2010, 2(4):257-265 ISSN No: 0975-7384 CODEN(USA): JCPRC5 Selective Extraction of 7-aminodeacetoxycephalosporanic acid and Cephalexin antibiotics with Aliquat-336 M M Bora North East Institute of Science and Technology, Jorhat -785006, Assam India (Council of Scientific & Industrial Research) ______________________________________________________________________________ ABSTRACT Thereactive extraction of 7-ADCA (aminodeacetoxycephalosporanic acid) and Cephalexin with Aliquat-336 (tricaprylmethyl ammonium chloride) dissolved in n-butyl acetate as solvent has been studied over a pH range of 5 to 8. The effect of such parameters as Aliquat-336 concentration, pH and 7-ADCA and Cephalexin concentrations on distribution coefficient in the reactive extraction system has been evaluated. It was found that distribution coefficient (Kd) increases with increases of the aqueous phase pH implying that the anionic forms of the dissociated 7-ADCA and cephalexin are amendable for reactive extraction. Furthermore inference of practical relevance that could be drawn from the present study is that 7-ADCA and cephalexin can be selectively extracted from an aqueous solution typical of that encountered in biosynthesis of 7-ADCA and cephelexin. Study on reextraction of 7-ADCA and cephalexin from the extracted phase was carried out using an aqueous solution of acetate was found to be pH dependent which also attribute to ionization behaviour of the beta-lactams. Such observation is considered important as re-extraction at an appropriate pH value is possible. Keywords: Distribution coefficient, 7-ADCA exchange, re-extraction, Aliquat-336. ______________________________________________________________________________ INTRODUCTION Reactive extraction in liquid membrane can provide an effective method for separation and purification of cephalosporin antibiotics from dilute solution [1, 2] such as that encountered in and Cephalexin antibiotics, Ion 257

2. M M Bora J. Chem. Pharm. Res., 2010, 2(4):257-265 _____________________________________________________________________________ fermentation broth. Accordingly, extensive studies have been performed to demonstrate the feasibility of liquid membrane system, which can provide enhanced separation potential [1, 3, 4]. We had reported in our previous papers the complementary studies on reactive extraction of 6- APA and 7-ACA with extractants such as secondary, tertiary and quaternary amines and it was found that Aliquat-336 was the best choice of carrier for specific cases [4-9]. The same carrier has been found to be effective for reactive extraction of amino acids, lactic acid [11-12] and other anionic species [14]. However, a systematic study on the equilibrium of reactive extraction has not been reported earlier in the literature. Our investigation on reactive extraction of cephalosporins with Aliquat-336 revealed that butyl acetate as the ideal solvent provides the highest extraction efficiency of 7-ACA [4]. Any probability of the non-identity of the organic phase can be ruled out because of the negligibly small aggregation of the lipophilic carrier in polar butyl acetate as the solvent. In this paper, we report a comprehensive study on distribution coefficient of 7-ADCA and Cephalexin with Aliquat-336 in butyl acetate as the solvent system and try to assess the selectivity from results obtained from single component extraction experiment. Biosynthesis of antibiotics has drawn special interest from the researchers throughout the world. In the bioconversion of 7-ADCA to cephalexin, purification and separation of cephalexin from the mixer of unconverted 7-ADCA is essential. Reactive extraction as such and in liquid membrane in particular, may be an effective method for this practically useful separation Theoretical considerations 7-ADCA and cephalexin are Zwitterionic molecules the pKa1 and pKa2 values being 2.95 and 4.78 for 7-ADCA and 2.56 and 5.88 for cephalexin. In the pH range between 2.95 and 4.78 for 7- ADCA and 2.56 and 5.88 for cephalexin, the zwitterions as a whole are predominant as evident from the dissociation behaviour shown in Figure 1 Figure 1 Dissociation behaviour of 7-ADCA and Cephalexin 258

3. M M Bora J. Chem. Pharm. Res., 2010, 2(4):257-265 _____________________________________________________________________________ The pH at which the amino group ionises keeps the carboxyl group in the COO- form. The anionic forms 7-ADCA and cephalexin at pH> pKa2 are amenable for ion-exchange with anionic exchanger such as Aliquat-336 (a carrier hereafter termed as QCl). Beta-lactam anion (P-) complexes with the carrier, QCl, dissolved in the organic solvent according to the following reaction. P- + QCl QP + Cl- (1) The anion exchange reaction takes place in the interface of the aqueous organic phase, the P- ion being extracted as a complex QP to the organic phase liberating Cl- into the aqueous phase. 7-ADCA (HP) and cephalexin (HP) molecule first dissociates in aqueous solution at pH > pKa2 to give the anion P- and a proton, H+ as follows HP H+ + P- (2) The dissociation equilibrium constant, K of this reaction is given by CH CP- Kp = -------------- (3) CHP The distribution coefficient (Kd) of HP is defined as CQP Vo Kd = -------------- (4) CHP Va Co-extraction of inorganic anion is equally important because is the biosynthesis media contains buffer anions such as phosphate, sulphate etc. Thus the co-extraction would effect the extraction of primary beta-lactam anion and the extent of the co-extraction should be quantified from process design, point of view. Since the co-extraction is relativity low at high pH value [4], it is not considered for analysis of extraction selectivity in the present work. EXPERIMENTAL SECTION Chemicals and reagents 7-ADCA and cephalexin the beta-lactam antibiotics used in the work were procured from Sigma Chemical Co (St Louis, Missouri, USA) and are of 99.9% purity. Aliquat-336 (Aldrich Milwakee, USA) were used as received Butyl acetate and other analytical grade buffer reagents were procured from BDH (Mumbai, India) Procudure Equilibriumexperiments The distribution coefficient experiments were conducted by using 10 ml each of aqueous solution of 7-ADCA and cephelexin and the organic solution of Aliquat-336 dissolved in butyl acetate in a stoppered 50 ml round- bottom flask which was placed over a magnetic plate and the 259

4. M M Bora J. Chem. Pharm. Res., 2010, 2(4):257-265 _____________________________________________________________________________ solution was agitated with a magnetic needle maintaining the temperature at 250C. The aqueous phase pH was maintained in the range of 5 to 8 by using. The method for preparation of the buffer solution is the same as reported in our previous papers [4, 5]. The time of equilibration for the extraction was 2 hours after optimization. After attainment of equilibrium, the two phases were allowed to settle, aqueous and organic phases were separated and aqueous of 0.1 ml aqueous solution was taken for analysis of beta-lactam content by using a UV-Visible Spectrophotometer (Shimadzu, Model 160 A) calibrated at the wave lengths of 265 and 262 nm for 7-ADCA and cephalexin, respectively. The distribution coefficient experiments were carried out in triplicate and the reproducibility was found to be ± 5%. The distribution coefficient was calculated as the ratio of beta-lactam content in organic to that in aqueous phase. Organic phase concentration was estimated from mass balance in the phase. For stripping, the aqueous solution taken was an acetate buffer of pH 4 made by using 10 MolL-1 acetate ion. In the organic phase, the initial concentration of solute-carrier complex (QP) and free carrier (QCl) was varied from 0.7 to 0.9 mM and 0.52 to 9.8 mM, respectively. The organic phase was loaded with beta-lactam ion by equilibrating equal volumes of the aqueous beta- lactam solution at appropriate pH and n-butyl acetate solution of Aliquat-336 (organic phase) under intense stirring in a glass vessel for two hours. The concentration, of the complex (QP) was varied by taking different concentrations of QCl during loading experiment and determine from a material balance. The stripping experiments were carried out at 250C by mixing 10 ml each of buffered aqueous solution and loaded organic solutions in equal volumes in a stopper glass flask of 50 ml capacity round- bottom flask placed over a magnetic plate with magnetic capsule. The aqueous phase pH of 7-ADCA and cephalexin was maintained at a value of 3 to 4 by using acetate buffer respectively. Samples of the aqueous phase were collected at equal interval of time and the concentration of the solute was determined by a UV-Visible Spectrophotometer calibrated at appropriate wave length [4]. RESULTS AND DISCUSSION The reactive extraction behaviour expressed in terms of distribution coefficients (Kd) of 7- ADCA and cephalexin as a function of aqueous phase pH and carrier concentrations are shows in Figures 2 and 3. The value of Kd increase with an increase of pH (upto pH 8), the extent of increase being higher for high range of carrier concentration such as observation of strong pH dependence of Kdat high carrier concentration was obtained also for the extraction of dl- phenylamine with Aliquat-336 6,10 and other beta-lactam reported by us elsewhere [4]. Different Kd values at identical values of pH for different beta-lactams provide further evidence on the effect of chemical nature of the beta-lactams on extraction efficiency of Aliquat-336 [4]. Furthermore, finite and significant difference in distribution coefficients at high and low pH values as is evident from Figure 2 and 3 implies that extraction can be carried out at high pH while stripping will take place at lower pH which is advantageous from practical point of view. 260

5. M M Bora J. Chem. Pharm. Res., 2010, 2(4):257-265 _____________________________________________________________________________ 3.5 3.0 2.5 2.0 Kd 1.5 1.0 0.5 0.0 5.0 5.5 6.0 6.5 7.0 7.5 8.0 pH Figure 2 Effect of pH on Kd at various carrier concentrations for extraction of 7-ADCA with Aliquat-336. CP = 1 mM ; Symbol, Aliquat –336 = ? , 10mM ; Ο ,5 mM ; ∆, 2 mM ; ∇,1 mM. 7 6 5 4 Kd 3 2 1 0 5.0 5.5 6.0 6.5 pH 7.0 7.5 8.0 Figure 3 Effect of pH on Kd at various carrier concentrations for extraction of cephalexin with Aliquat-336. CP = 1 mM ; Symbol, Aliquat –336 = ? , 10mM; Ο , 5 mM ; ∆, 2 mM ; ∇,1 mM The effect of beta-lactam (solute) concentration (at constant pH) of aqueous phase on Kd was studied in the solute concentration range of 0.5 to 1.5 mM at a carrier concentration of 1 mM. As is evident from Figures 4 and 5, Kddecreases with equilibrium solute concentration. This dependence is identical for various carriers studied in our previous works [3], the highest Kd value under otherwise identical experimental conditions was observed with Aliquat-336 from which it may be inferred that the ion-exchange extraction with such lipophilic carrier will perhaps be the best choice for reactive extraction of cephalosporin antibiotics. The Figures 4 and 5 also show that Kdvalue decreasewith an increase of solute concentration in general, which may be attributed to a shift (decrease) in the equilibrium pH value probably as a result of the reduction in the mole fraction of the extractable anion. This probability is further supported by the observation of a lower Kd value at low pH, particularly at low equilibrium concentration of 261

6. M M Bora J. Chem. Pharm. Res., 2010, 2(4):257-265 _____________________________________________________________________________ the solute. This observed effect of solute concentration and pH on Kd for extraction of all the beta-lactams with Aliquat-336 appears to be identical. However, the lowering of Kd values with decreasing pH is more pronounced in the lower range of solute concentration. 1.4 1.2 1.0 0.8 Kd 0.6 0.4 0.2 0.0 0.4 0.6 0.8 1.0 1.2 1.4 1.6 CP (mM) Figure 4 Effect of solute concentration on Kd for extraction of 7-ADCA with Aliquat-336 at various aqueous phase pH. CQCl = 1 mM ; Symbol, pH= ? , 8 ; Ο, 7 ; ∆ , 6 ; ∇, 5 1.8 1.6 1.4 1.2 1.0 Kd 0.8 0.6 0.4 0.2 0.0 0.4 0.6 0.8 1.0 1.2 1.4 1.6 CP(mM) Figure 5 Effect of solute concentration on Kd for extraction of cephalexin with Aliquat-336 at various aqueous phase pH. CQCl = 1 mM ; Symbol, pH= ? 8 ; Ο, 7 ; ∆ , 6 ; ∇, 5. It is thought worth while to assess the selectivity of extraction as expressed be the following relation Kd Cephelexin Selectivity (s) = ----------------- (5) Kd 7-ADCA Where Kd Cephelexin and Kd 7-ADCA represent asdistribution coefficient of Cephelexin and 7-ADCA respectively and were determined in dependant by using symmetric sample. Figure 6 is plot of 262

7. M M Bora J. Chem. Pharm. Res., 2010, 2(4):257-265 _____________________________________________________________________________ Selectivity (s) vs pH for various experimental conditions. From Figure 6, it is apparent that the reactive extraction exhibits selectivity for cephelexin by an order of magnitude. This implies that from a binary mixture of Cephalexin and 7-ADCA like that encountered in the biosynthetic media in cephelexin production, selective separation of cephelexin is possible at a pH value above the pKa2 of cephelexin. Thus reactive extraction using Aliquat-336 can be exploited in a liquid membrane system for practical purpose. 2.4 1 2.2 Selectivity for Cephelexin 2.0 2 1.8 1.6 3 4 1.4 1.2 1.0 0.8 0.6 4.5 5.0 5.5 6.0 6.5 pH 7.0 7.5 8.0 8.5 Figure 6 Selectivity for cephelexin vs pH at various carrier concentrations. Aliquat –336 concentration in plot , 1 =? , 10mM; 2 = Ο ,5 mM ; 3 = ∆, 2 mM ; 4= ∇,1 mM. It appears that the experimental Cps versus time (t) profile is fairly well for 7-ADCA and cephalexin shown in figure 7. The loaded organic phase contains appreciable proportion of free QCl which has certain surface active properties and is likely to take part in another anion exchange reaction with buffer anion of the stripping aqueous solution. The re-extraction of cephalexin and 7-ADCA are 81% and 65.5% respectively. 9 Symbol Beta-lactam 7-ADCA Cephalexin 8 7 6 Cps/CQP 5 4 3 2 0 20 40 60 80 100 120 140 t Minute Figure 7 Re-extraction of 7-ADCA and Cephalexin 263

8. M M Bora J. Chem. Pharm. Res., 2010, 2(4):257-265 _____________________________________________________________________________ CONCLUSION A liquid-liquid ion exchange a mechanism can be exploited for reactive extraction of 7-ADCA and cephalexin from aqueous solution using Aliquat-336 and butyl acetate as the carrier and solvent, respectively. The partition coefficients are influenced by the aqueous phase pH, carrier and solute concentration. The reactive extraction system can provide an attractive method for selectivity separation of cephelexin from a binary mixture of cephelexin and 7-ADCA. An additional ion-exchange reaction between free QCl and buffer anion is expected to facilitate the stripping process and the role of additional Cl− concentration seems to be insignificant. Acknowledgement: Authors would like to Dr N N Dutta, Head, and Department of Chemical Engineering, NEIST, Jorhat and Dr P G Rao, Director, NEIST, Jorhat. Nomenclature 6-APA 6-aminopenicillinic acid 7-ACA 7-aminocephalosporinic acid 7-ADCA 7-aminodeacetoxycephalosporanic acid C concentration H+ proton, mol/m3 HP 7-ADCA and cephalexin anion, mM K dissociation equilibrium constant Kd distribution coefficient Kd Cephelexin distribution coefficient of Cephelexin Kd 7-ADCA distribution coefficient of 7-ADCA P- 7-ADCA and cephalexin anion, mM QCl carrier, Aliquat-336, mM QP solute-carrier complex, mM S Selectivity V Volume Subscript/superscript a aqueous phase o organic phase REFERENCES [1]A C Ghosh, M M Bora, R. K Mathur N N Dutta, Bioseparation, (1996) 6, 91- 105. [2]A C Ghosh, R K Mathur, N N Dutta, Advances in Biochem. Eng./Biotech.,(1997), 56,111- 146. [3]M M Bora, K G Bhattacharya N N Dutta, J.Chem.Eng. Data, (1998), 43,318-324. [4]M M Bora, N N Dutta and K G Bhattacharya, Chem.Eng. Commun. (2000), 179, 15-34. [5]M. M. Bora, A. C. Ghosh, N. N.Dutta, Can. J. Chem. Engg., (1997) , 75, 520- 526. [6]T Hano, M Matsumoto, T Ohtake F Hori, J. Chem. Eng. Japan., (1992), 25,293-296. [7]K H Lee, W K Lee, Separ. Sci. Technol, (1994), 29 (5), 601-625. [8]S T Yang , S A White, S T Hsu ,Ind. Eng. Chem. Res., (1991), 30, 1335-342, 264

9. M M Bora J. Chem. Pharm. Res., 2010, 2(4):257-265 _____________________________________________________________________________ [9]T Hano, T Ohtake, M Matsumoto, D Kitayama, F Hori and F Nakashio, J. Chem. Eng. Japan, (1991), 24, 20-24, [10]R Haensel, W Halwach K Schugerl, Chem. Eng. Sci.,(1986), 41(7), 1811- 1815. [11]B Galan , A M Uritaga , A I Alonso , J A Irabien M I Ortiz , Ind. Eng. Chem. Res., (1994), 33,1765-1770, [12]M. Reschke, K. Schurgerl The chem Eng J., (1984), 29 B,25-29. [13]S H Asai, Y Furichi Nakamura, J. Chem. Japan, (1991), 24, 53-658. 265